AnarchyeL
28-08-2005, 23:02
Talk about doing science backwards. You are doing the same right now. You have decided that because you dislike the source you will refuse to consider the data.
On the contrary, I dislike the source because of how he treats the data. He picks and chooses his data in order to make misleading claims.
I believe that is the title of Mr. Farrell's book [Why Men Earn More]... So far he's the only one to address it directly. Most others seem to focus on how much more. I've seen precious few that don't blame it on a mysterious 'glass ceiling' or 'conspiracy'. Mr. Farrell is one of the few to address more tangible possibilities.
That is not true at all. Social scientists approach the problem from a variety of perspectives. Among others, they have tried an experimental approach. For instance, it has been found that when job applicants are interviewed "blind" -- so that the interviewer cannot identify gender -- women are hired in greater proportions and at higher salaries. The experimental result supports the prejudicial theory. More importantly, they do not use distorted numbers to make their case.
To quote you; "That would be very interesting indeed!! If it were true... which it is not." You are now just making stuff up.
It is common, in good writing, to "introduce" one's conclusions before making an argument. I believe I backed up this claim in the paragraph that followed.
Mr. Farrell makes no such claim. The title of his book alone should be enough to indicate that.
Yes, it makes a very nice cover for his argument to disguise it as theory. However, the essence of what he is saying is that "there is no problem." "Men and women are paid equally for equal work." But this is not what the numbers show. Men make more, in general, regardless. Men do not, as his title suggests, simply "earn" more.
Ah, so you endorse 'selective sampling'. Throwing out the 'exceptions' as 'anomolous' data that contradicts the 'norm'.
No, now you are just making things up. Being a scientist, I know that all the available data must be analyzed to identify a trend. It is Dr. Farrell who selects the data he wants in order to prove a point. The idea, for a scientist, is not to "throw out" any data. The idea is that when all the data is taken into consideration, general trends appear. It is the trends that are statistically significant... Specific measurements, on specific industries, are beside the point (unless that is what you set out to study, and the phenomena about which you made predictions). Again, if Dr. Farrell had predicted, for some reason, that when data on statisticians were analyzed it would reveal that women make more than men, his conclusion would be scientific. However, with the way he has done it, not only will those highly paid female statisticians disagree with him, but so will their less well-paid male colleagues. It is simply not science.
Again, I think the title of the book indicates that those results are an exception and not the rule quite well. Do you really need me to remind you what it is? You are a scientist - refer to your notes. I'm sure you wrote it down somewhere...
To reiterate, his title is simply misleading. The scientist concludes that men do not earn more, they simply make more.
Umm, hmm. What age group represents the largest portion of the workforce. Over 65? Nope. Under 18? Nope. 40-64?
If they really represented the largest portion of the workforce, then Dr. Farrell would not need to zero-in on them to make his point. Statistically, the largest portion should outweigh the rest, right? So why does he not just say, "statistically, single college-educated women make more than similar men?" Simple: he can't. When the entire workforce is taken into account, the fact is that men make more than women.
Perhaps more importantly, people who have never married in the 40-64 age-group make up a tiny fraction of the overall workforce. He zeroes-in even more to find his exception to the rule.
On a different data set, a different group of women might make more than comparable men. Perhaps now it is high-schoolers, or never-marrieds with advanced degrees aged 27-45, or post-retirees. And Farrell could never predict which it would be, because these are fluctuations, not trends. When you show me a prediction he has made and tested, we can talk about his scientific work. But not before.
(Of course, it could also be that the women he picks out consistently make more than comparable men... we just don't know, because he has not tested the hypothesis. Even if it turns out to be true, however, this group remains a pitifully small fraction of the workforce. How many people have never been married by age 64?) EDIT: According to the Census Bureau, only about 3% of men and 2% of women have never been married by age 40. By age 65? Less than one-half of one percent each.
You are so wrong here I don't even know where to start. Data is data.
Exactly. If only Dr. Farrell thought so.
Your hypothesis prior to collecting it does not have any relevance to it.
Tell that to a scientist. To do science, one does not simply study survey data looking for correlations, then claim that one has found something. ("Oh, look!! It seems that blondes live longer than brunettes!!") When dealing with a large number of variables, meaningless correlations pop up all the time. In order to test whether or not a correlation one has found actually exists, one has to test it on new data... because it may just be a quirk of this particular population.
This is quantitative methods 101... first year, grad school (and sometimes earlier).
You didn't question the 'abberation' when it gave you the concluison that you sought.
That is because the scientists who conclude that men make more than women (regardless of other factors) include all the data in their analysis.
When someone else took a look at the same data and illustrated a conclusion that you disliked you suddenly decide the data is invalid??
You really need to reread what I wrote. The data is valid. Isolating it from the rest of the data is not.
Buried deep in this hypocritical rant is your admission that that I had correctly pointed out that you were wrong in your presumption that porn is the only industry that women earn more than men.
Actually, you have yet to provide valid scientific evidence of this. You only have meaningless, untested correlations to show me.
On the contrary, I dislike the source because of how he treats the data. He picks and chooses his data in order to make misleading claims.
I believe that is the title of Mr. Farrell's book [Why Men Earn More]... So far he's the only one to address it directly. Most others seem to focus on how much more. I've seen precious few that don't blame it on a mysterious 'glass ceiling' or 'conspiracy'. Mr. Farrell is one of the few to address more tangible possibilities.
That is not true at all. Social scientists approach the problem from a variety of perspectives. Among others, they have tried an experimental approach. For instance, it has been found that when job applicants are interviewed "blind" -- so that the interviewer cannot identify gender -- women are hired in greater proportions and at higher salaries. The experimental result supports the prejudicial theory. More importantly, they do not use distorted numbers to make their case.
To quote you; "That would be very interesting indeed!! If it were true... which it is not." You are now just making stuff up.
It is common, in good writing, to "introduce" one's conclusions before making an argument. I believe I backed up this claim in the paragraph that followed.
Mr. Farrell makes no such claim. The title of his book alone should be enough to indicate that.
Yes, it makes a very nice cover for his argument to disguise it as theory. However, the essence of what he is saying is that "there is no problem." "Men and women are paid equally for equal work." But this is not what the numbers show. Men make more, in general, regardless. Men do not, as his title suggests, simply "earn" more.
Ah, so you endorse 'selective sampling'. Throwing out the 'exceptions' as 'anomolous' data that contradicts the 'norm'.
No, now you are just making things up. Being a scientist, I know that all the available data must be analyzed to identify a trend. It is Dr. Farrell who selects the data he wants in order to prove a point. The idea, for a scientist, is not to "throw out" any data. The idea is that when all the data is taken into consideration, general trends appear. It is the trends that are statistically significant... Specific measurements, on specific industries, are beside the point (unless that is what you set out to study, and the phenomena about which you made predictions). Again, if Dr. Farrell had predicted, for some reason, that when data on statisticians were analyzed it would reveal that women make more than men, his conclusion would be scientific. However, with the way he has done it, not only will those highly paid female statisticians disagree with him, but so will their less well-paid male colleagues. It is simply not science.
Again, I think the title of the book indicates that those results are an exception and not the rule quite well. Do you really need me to remind you what it is? You are a scientist - refer to your notes. I'm sure you wrote it down somewhere...
To reiterate, his title is simply misleading. The scientist concludes that men do not earn more, they simply make more.
Umm, hmm. What age group represents the largest portion of the workforce. Over 65? Nope. Under 18? Nope. 40-64?
If they really represented the largest portion of the workforce, then Dr. Farrell would not need to zero-in on them to make his point. Statistically, the largest portion should outweigh the rest, right? So why does he not just say, "statistically, single college-educated women make more than similar men?" Simple: he can't. When the entire workforce is taken into account, the fact is that men make more than women.
Perhaps more importantly, people who have never married in the 40-64 age-group make up a tiny fraction of the overall workforce. He zeroes-in even more to find his exception to the rule.
On a different data set, a different group of women might make more than comparable men. Perhaps now it is high-schoolers, or never-marrieds with advanced degrees aged 27-45, or post-retirees. And Farrell could never predict which it would be, because these are fluctuations, not trends. When you show me a prediction he has made and tested, we can talk about his scientific work. But not before.
(Of course, it could also be that the women he picks out consistently make more than comparable men... we just don't know, because he has not tested the hypothesis. Even if it turns out to be true, however, this group remains a pitifully small fraction of the workforce. How many people have never been married by age 64?) EDIT: According to the Census Bureau, only about 3% of men and 2% of women have never been married by age 40. By age 65? Less than one-half of one percent each.
You are so wrong here I don't even know where to start. Data is data.
Exactly. If only Dr. Farrell thought so.
Your hypothesis prior to collecting it does not have any relevance to it.
Tell that to a scientist. To do science, one does not simply study survey data looking for correlations, then claim that one has found something. ("Oh, look!! It seems that blondes live longer than brunettes!!") When dealing with a large number of variables, meaningless correlations pop up all the time. In order to test whether or not a correlation one has found actually exists, one has to test it on new data... because it may just be a quirk of this particular population.
This is quantitative methods 101... first year, grad school (and sometimes earlier).
You didn't question the 'abberation' when it gave you the concluison that you sought.
That is because the scientists who conclude that men make more than women (regardless of other factors) include all the data in their analysis.
When someone else took a look at the same data and illustrated a conclusion that you disliked you suddenly decide the data is invalid??
You really need to reread what I wrote. The data is valid. Isolating it from the rest of the data is not.
Buried deep in this hypocritical rant is your admission that that I had correctly pointed out that you were wrong in your presumption that porn is the only industry that women earn more than men.
Actually, you have yet to provide valid scientific evidence of this. You only have meaningless, untested correlations to show me.